各位大神好，我在用MPLUS处理追踪数据的时候遇到一些困难，想问问如何解决
有三次数据，采用多重插补法处理缺失值，然后建立了潜变量混合增长模型，然后死循环了，下面是MPLUS代码
TITLE: Missing Data;
DATA: FILE IS 3depress.dat;
VARIABLE:
NAME are gender f1-f18 b1-b18 c1-c18;
USEVARIABLES =gender f1-f18 b1-b18 c1-c18;
MISSING are ALL (9999);
DATA IMPUTATION:
IMPUTE = b1-b18 c1-c18;
  NDATASETS = 50;
  save = depressimp*.dat;
ANALYSIS:
   type = basic;
  bseed = 79566;
  chains = 1;
TITLE: Summary Statistics with Multiply Imputed Data;
DATA: FILE IS depressimplist.dat;
      TYPE = imputation;
VARIABLE:
NAME are gender f1-f18 b1-b18 c1-c18;!三次追踪题项
USEVARIABLES = gender f1-f18 b1-b18 c1-c18 DFOUR DFIVE DSIX;
               class=c(2);
ANALYSIS:
  ESTIMATOR = ML;
  INFORMATION = observed;
  TYPE=MIXTURE;
  START=500 50;
  processor=4;        
DEFINE:
DFOUR=MEAN(f1-f18);!第一次均值
DFIVE=MEAN(b1-b18);!第二次均值
DSIX=MEAN(c1-c18);!第三次均值         
MODEL:
  %OVERALL%
  i s|DFOUR@0 DFIVE@1 DSIX@2;
  OUTPUT:TECH11 TECH14 TECH4 ;
  SAVEDATA:FILE=DEPRESS2.dat;
  SAVE=cprob;
  plot:TYPE is PLOT3;
直接删除缺失值跑数据，提示矩阵非正定，已经尝试过将斜率方差固定为0，但是得到的分组人数差距太大，一组400多，一组10，我想问问是不是模型不成立，或者数据有问题
WARNING:  THE LATENT VARIABLE COVARIANCE MATRIX (PSI)  IN CLASS 1  IS NOT POSITIVE DEFINITE.  THIS COULD INDICATE A NEGATIVE VARIANCE/RESIDUAL VARIANCE FOR A LATENT VARIABLE, A CORRELATION GREATER OR EQUAL
TO ONE BETWEEN TWO LATENT VARIABLES, OR A LINEAR DEPENDENCY AMONG MORE  THAN TWO LATENT VARIABLES.  CHECK THE TECH4 OUTPUT FOR MORE INFORMATION.PROBLEM INVOLVING VARIABLE S.
这是TECH4输出的结果

           ESTIMATED MEANS FOR THE LATENT VARIABLES
              I             S
              ________      ________
1              1.645         0.008


           ESTIMATED COVARIANCE MATRIX FOR THE LATENT VARIABLES
              I             S
              ________      ________
I              0.003
S              0.009        -0.005


           ESTIMATED CORRELATION MATRIX FOR THE LATENT VARIABLES
              I             S
              ________      ________
I              1.000
S            999.000       999.000